P/F Ratio May Be a Marker of Potential Intraoperative Ventilator-Induced Lung Injury

Caroline Helwick

October 23, 2009

October 23, 2009 (New Orleans, Louisiana) — In patients without existing lung injury, ventilator-induced lung injury (VILI) occurs in 24% after 5 days of ventilation. The ratio of arterial oxygen concentration to the fraction of inspired oxygen (P/F ratio) is 1 of 3 criteria for lung injury and might be a marker of potential VILI in patients undergoing general anesthesia, University of Michigan investigators reported here at the American Society of Anesthesiologists 2009 Annual Meeting.

"It is known that inflammatory markers associated with lung injury are present as early as 5 hours in patients under general anesthesia, but there is little evidence regarding the development of lung injury over the course of anesthesia, or its risk factors," said lead author Ravi Tripathi, MD, a critical care fellow at the University of Michigan School of Medicine, Ann Arbor.

"We asked if, over the course of anesthesia delivery, oxygenation (using P/F ratio as a marker) may decrease," he said. "The results of our study suggested that there is a potential for VILI during routine intraoperative ventilation over time."

James Blum, MD, a senior investigator, explained further: "The P/F ratio is 1 of 3 things we need to detect lung injury, but it is the only one that is occurring during the case. We can possibly isolate this at-risk group, based on P/F ratio, ask whether they are getting worse and whether this is due to VILI, and decide whether to use a strategy that might protect them."

This was a retrospective study of all adults undergoing general anesthesia over a 3-year period, excluding cardiac and thoracic procedures and patients classified as ASA 4 or 5 (i.e., with serious/lifethreatening systemic disease). Patients with a P/F ratio of less than 300 at baseline were excluded, and those with a subsequent P/F ratio of less than 300 during anesthesia were identified as having potential VILI.

The data were based on 9633 blood gas measurements for 2946 cases. The mean age of the patients was 57 years, mean case length was 336 minutes, and mean initial P/F ratio was 432. ASA classification was 3/3E (60.5%), ASA 2/2E (36.9%), or 1/1E (2.6%).

Of the 2946 patients, 1145 (27.1%) were identified as potentially having VILI (P/F ratio < 300), and 3 independent predictors were identified in the study: age, case length, and ASA classification.

"We found that risk was increased about 2-fold for ASA 3 classification versus ASA 1 or 2, for case length longer than 11 hours, and for age over 60," Dr. Tripathi said.

The odds ratios for having a P/F ratio of less than 300 were 1.61 for age older than 60 years (95% confidence interval [CI], 1.35 - 1.98). For ASA classification, 29.5% of potential VILI patients had an ASA of 3, 23.9% had an ASA of 2, and 16% had an ASA of 1. The odds ratio for ASA 3 was 1.97 (95% CI, 1.63 - 2.38).

The odds ratio for a case length longer than 6 hours was 1.22 (95% CI, 1.02 - 1.46), and was 2.02 (95% CI, 1.78 - 2.29) for cases longer than 11 hours.

"When we evaluated these 3 variables in binary logistic regression to potential VILI or a minimum P/F ratio of less than 300, all 3 remained independent predictors of a potential VILI," Dr. Tripathi said.

The study also showed that patients who began the study at a P/F ratio of approximately 300 to 350 had a "downward trend for oxygenation," whereas those with higher values, around 400 to 450, "improved over the course of the anesthetic," he added. "We would like to learn what makes these 2 groups different."

The next step, having identified a population that appears to be at high risk for VILI, is to refine the association between risk factors, P/F ratios, and postoperative outcomes, he said.

Nicholas Sadovnikoff, MD, from Brigham and Women's Hospital in Boston, Massachusetts, who moderated the poster session, said that the data are preliminary and hypothesis-generating for intraoperative pulmonary deterioration. "Ventilator management is an active topic, as it is important in adult respiratory distress syndrome and acute lung injury," he said. "The study is a building block to a prospective type of analysis."

Dr. Tripathi and Dr. Blum have disclosed no relevant financial relationships.

American Society of Anesthesiologists 2009 Annual Meeting: Abstract A1448. Presented October 21, 2009.